Where Are We, What Are We, Why Are We? Cover
This is an extract from Chapter 3 of

Where Are We, What Are We,
Why Are We?

And Why Do We Want To Know?

Buy the book at your local Amazon:


Anthropomorphic Projection

Face Recognition Software

Why do we see the shapes of people and animals where there aren’t any: in inkblots, in clouds, in cliffs, in taps?
It could be because our brains are wired to prioritise human and animal forms in the visual environment. Basically, we’re obsessed with such forms. Our brains devote a huge amount of circuitry to the recognition of them, with a special emphasis being put on human faces. Here’s an explanation of how our brains do it, followed by the reason why they bother.
Firstly, it’s important to realise that the brain doesn’t see the world around it simply as though the scene was projected onto a cinema screen on the inside of your skull. Before a scene can be observed “in your head” it has to be broken down into a number of different components for processing, and these components then have to be recombined into the meaningful form that we call “an image”. Amongst other things, the scene is broken down into its different colours – red, green and blue – in a way that’s analogous to the manner in which a television image or magazine photograph is broken down into tiny dots of primary colours (which are too small to be noticed individually when we look at them, but which when seen collectively give the impression of a continuous full colour image). However, unlike TV and magazine images, the image that we see with our eyes is broken down not only into separate colour components but into other components too. It is, rather incredibly, deconstructed into component parts such as horizontal lines, vertical lines, circles and so on. Each of these component parts is sent to a separate area of the brain for processing, with the different components of the scene only merging again when they are unified into what you perceive as the image.
The degree to which the scene that constitutes the outside world is split into separate components for processing is quite surprising. Not only are horizontal lines sent to a different part of the brain to vertical lines, but lines that are at 30, 45, 60 and so on go to their own individual areas too, and even movement is processed in its own discrete areas of the brain by its own dedicated brain cells.
The separate areas of the brain that are used for processing different types of visual information can be detected by the use of brain scanning techniques such as functional magnetic resonance imaging (fMRI). This technique works by probing the brain with the use of powerful magnetic fields. Areas of the brain that are busy processing data have higher blood flow than other areas, and this increased flow is detected by the scanner and shows up as light areas (Figure 20 shows an artist’s impression of such a scan).
The resulting image has something of the Rorschach test about it, don’t you think? Perhaps you can deduce the personality of the brain’s owner from it.

mri scan fmri scan
Figure 20:  An fMRI brain scan, with the active areas of the brain showing up lighter than the rest

When the person whose brain is being scanned is looking at a horizontal line a particular part of the brain lights up, and when looking at a vertical line a different part lights up. And so on.
If a person suffers brain damage to an area of the brain that deals with a particular type of visual information the ability to process that information may be impaired. For instance if the region that deals with horizontal lines is damaged, horizontal lines may drop out of the final image, while vertical ones and angled ones are still seen with no problem. Thus a person with this condition may be unable to see horizontal objects such as bookshelves but be able to see the books that are resting on them because their spines are vertical.
If you have trouble visualising this selective blindness and you think that it sounds too bizarre to be true, bear in mind that each of your own eyes is selectively blind already, with its own blind spot – an area that is blind to everything – just off to one side of the centre of vision.
This blind spot is caused because there is a portion of the retina at the back of your eye that has no light detecting sensors, due to the fact that the area is occupied by the “cable” that takes the visual information away from your eye and into your brain.
If you’re not familiar with the blind spot, here is how to notice its presence – by staring at Figure 21.

Blind spot exercise
Figure 21: Find your blind spot by making the rabbit disappear.

 Stare at the hat using only your right eye while holding the image at about half an arm’s length. Move closer or further until the rabbit vanishes
Close your left eye and look at the hat in the figure using only your right eye. Slowly move the page closer to or further from your eye, at a distance somewhere in the region of half an arm’s length, while you stare at the hat. At some point you’ll notice that the rabbit vanishes, as though by magic. This is because the part of your retina that should be seeing the rabbit is occupied by your blind spot.
As well as having areas that process basic components of images such as angled lines and colours, the brain has areas that process more complex forms that are particularly significant. Here in Figure 22 is a shape that you’ll process in such an area.

face recognition algorythm
Figure 22:  This image is nothing more than an ellipse and two dots

Yes, it’s a human face.
But of course, it isn’t a human face at all: it’s just an ellipse with a couple of dots on it. The brain overrides this obvious fact, with the consequence that it takes a real effort not to see the image as a face. The brain seems to contain something similar to a standardized face template, and any object or form that exhibits the basic ingredients of a face will have its shape shunted off to the face recognition part of the brain. Notice how the shapes in Figure 23 on the next page aren’t automatically interpreted as faces, even though their components are exactly the same as those in Figure 22 apart from their positions.

face regognition in brain
Figure 23:  These images are nothing more than ellipses and dots too

When it comes to seeing human faces, the “face recognition software” in our brains is so powerful that not only can we recognise the general shape of faces instantly, but we can also differentiate between millions of different individual real faces when we come across them. This ability is quite amazing, and is extremely useful in the modern world with its teaming millions-too-many people. Its power is quite intriguing however, due to the fact that when this ability was first developed it must have seemed almost like overkill, as in pre-modern times you’d be lucky if you ever had the need to differentiate between a few hundred individual faces at most.
If the part of the brain that is used for analysing individual faces is damaged it becomes impossible to recognise people from their facial appearance. Sufferers can’t recognise their own friends and families, or sometimes even their own reflections. This condition is known as face blindness, or prosopagnosia.
As a result of the mass of face recognition software that it comes bundled with, the brain is rather over-enthusiastic in seeking out faces, so it frequently sees them where there aren’t any. In other words, it makes a lot of mistakes.
I realise that “making mistakes” isn’t as poetic or romantic a reason for seeing faces everywhere as some readers may wish for. However it’s because of its very mundanity and blandness that I suspect that it’s probably right. Mundanity generally trumps poeticalness on the principle that if an explanation for anything has an excessively positive or pleasing feel to it one should be suspicious of it on the grounds that it’s got the dubious whiff of wishful thinking about it.

The Plus Side of False Positives

The type of mistake that the brain makes when it sees a face where there isn’t one is called the generation of a false positive. A false positive occurs when you think that you’ve made a positive identification of something, but you were mistaken.
Generating false positives is a different type of mistake to the type that you make as a result of carelessness, such as tripping over your shoelaces or forgetting your partner’s birthday. False positives are often a necessary consequence of the need to provide a safety margin when making quick identifications.
Here’s a typical example of the generation of a false positive, taken from the world of modern domestic gadgetry.

To read on, please think about purchasing the book.
It's available from Amazon here: